PhytoScience - Article

NSAID - Drug-Induced Enteropathy

Mayo Clin Proc. 1995;70:55-61 © 1995 Mayo Foundation for Medical Education and Research

Case Report

Nonsteroidal Anti-Inflammatory Drug-Induced Enteropathy: Case Discussion and Review of the Literature

AUL Y. KWO, M.D.; WILLIAM J. TREMAINE, M.D.

From the Division of Gastroenterology and Internal Medicine, Mayo Clinic Rochester, Rochester, Minnesota.

Address reprint requests to Dr. W. J. Tremaine, Division of Gastroenterology, Mayo Clinic Rochester, 200 First Street SW, Rochester, MN 55905.

Abstract

The adverse effects of nonsteroidal anti-inflammatory drugs (NSAIDs) on the upper gastrointestinal tract are well described. Evidence also shows that NSAIDs can be harmful to the small intestine. The use of NSAIDs has been associated with small intestinal strictures, ulcerations, perforations, diarrhea, and villous atrophy. Herein we present a case of NSAID-induced enteropathy with multiple diaphragm-like strictures that involved the distal 35 cm of ileum and review the literature of other cases of NSAID-induced enteropathy in which biopsy specimens were obtained for histologic analysis to rule out other causes. The prevalence of NSAID-induced enteropathy is unknown. Diagnosis can be made by endoscopy or at abdominal exploration. The role of radionuclide scans for diagnosis remains unclear. The pathogenesis is likely multifactorial. Mucosal diaphragms may be specific for NSAID-related disease. Treatment options for NSAID-induced enteropathy are discussed.

(Mayo Clin Proc 1995;70:55-61)

GI=gastrointestinal tract; NSAIDs=nonsteroidal anti-inflammatory drugs; RBCs=red blood cells 

Nonsteroidal anti-inflammatory drugs (NSAIDs) are useful in the treatment of osteoarthritis, musculoskeletal pain, rheumatoid arthritis, and other inflammatory conditions;1 however, they cause increased mucosal damage to the upper gastrointestinal (GI) tract, including an increased incidence of gastric and duodenal ulcers.2,3 During the past decade, the deleterious effects of NSAIDs on the small intestine and colon have also been recognized4-6—changes in permeability, inflammation, and fibrosis with stenosis.7 Although the colon can be evaluated with colonoscopy, endoscopic examination of the jejunum and ileum with video enteroscopy is technically difficult and not available at most medical facilities. Two techniques have been advocated for quantification of inflammation in NSAID-induced enteropathy: the 111In scan and urinary excretion of 51Cr. To date, no prospective studies have shown these methods to correlate with histologic changes.8 Herein we present a case of histologically confirmed NSAID-induced enteropathy and review the literature of other such cases in which biopsy specimens were obtained for histologic examination.

REPORT OF CASE

A 37-year-old woman was hospitalized because of nausea, vomiting, large-volume diarrhea, and hypokalemia. Three years before the current admission, she underwent antrectomy, vagotomy, and gastroduodenostomy with a Billroth I anastomosis because of a nonhealing gastric ulcer. Two years previously, malaise, diarrhea, and persistent iron deficiency anemia developed; at that time, the hemoglobin concentration was 6.0 g/dL, and the ferritin level was 3 μg/L. She had been taking 800 mg of ibuprofen three times a day for pelvic pain for 2 years. A stomach roentgenogram with use of barium, esophagogastroduodenoscopy, colonoscopy, and biopsy of the small bowel showed normal findings. Results of a duodenal aspirate that was examined for Giardia and cultured for bacteria were negative. Because of recurrent anemia, she received a total of 6 U of packed red blood cells (RBCs) for 2 months, and ferrous sulfate was administered orally. Repeated colonoscopy demonstrated a nonbleeding 4-mm ulcer on the ileocecal valve. A biopsy specimen of the ulcer was cauterized, and the histologic examination revealed nonspecific ulceration. In addition, findings on a small bowel barium study, enteroclysis, and peroral small bowel biopsies were normal. Nonetheless, the anemia and diarrhea persisted.

The patient continued to take ibuprofen primarily for pelvic pain until hospitalization. On admission, findings on physical examination were unremarkable except for pallor. Laboratory studies revealed hypokalemia (serum potassium, 1.9 mEq/L) and iron deficiency anemia (hemoglobin, 6.0 g/dL); the serum ferritin level was 3 μg/L, and the stool quantification of blood (HemoQuant) was 7.8 mg total hemoglo-bin/g of feces (normal, less than 4). Visceral angiography demonstrated angiodysplasia in the cecum and ascending and proximal transverse colon. At laparotomy, multiple strictures in the distal ileum were palpated. The terminal 35-cm ileum was resected along with the right colon, and an ileotransverse colonic anastomosis was performed. Pathologic examination demonstrated seven diaphragm-like 1-cm strictures in the ileum with shallow circumferential ulcers in the area of strictures. Microscopic examination showed nonspecific ulceration. Postoperatively, the patient was instructed to discontinue use of ibuprofen. The anemia resolved, and she had no further anemia at 3-year follow-up.

LITERATURE REVIEW OF NSAID-INDUCED ENTEROPATHY

Twenty-five patients with NSAID-induced enteropathy have been described in the literature9-25 (Table 1). These patients, whose ages ranged from 40 to 78 years (mean, 66), were taking many types of NSAIDs for inflammatory conditions such as ankylosing spondylitis, osteoarthritis, and rheumatoid arthritis. The clinical manifestations were GI bleeding, obstruction, or diarrhea. Of the 21 patients in whom gender was reported, 14 were female, and 7 were male. Of the 16 patients in whom duration of NS AID use was reported, 11 had been taking NSAIDs for more than 1 year.9,10,12,15-17,23,24 An autopsy study also demonstrated a higher frequency of nonspecific ulcerations and erosions in patients who had taken NSAIDs for more than 6 months in comparison with control patients who had taken no NSAIDs.26 Nevertheless, five patients have, been described with NSAID-induced enteropathy after use of medication for less than 4 weeks.11,13,14,21,25 Moreover, the route of administration of NSAIDs does not seem to affect the development of enteropathy. For example, use of Osmosin, a now discontinued indomethacin rectal suppository, was associated with ileal perforations;14,21 14 of the 25 patients with NSAID-induced enteropathy were taking either indomethacin or piroxicam preparations. Of these 25 patients, 10 had intestinal obstruction, 10 had ulceration and anemia, and 3 had intestinal perforation (Table 1). In a recent autopsy series, the cause of death in three patients who had received long-term NS AID therapy was undiagnosed small bowel perforations; no such lesions were noted in the control group.26 Of 18 patients in whom histopathologic findings were reported, small intestinal strictures and ulcerations were noted in 12, villous atrophy in 3, and intestinal perforations in 3. Of the 25 patients, 21 underwent intestinal resection; in 4, the only treatment was discontinuation of use of NSAIDs. Of the 23 patients in whom follow-up was reported, 3 died, 2 of intestinal perforation.

In a retrospective study of 268 patients who were hospitalized because of small and large bowel perforations or hemorrhage, those who were taking NSAIDs were twice as likely to have one of these complications develop in comparison with control patients.27 Perforations were noted intraoperatively or at autopsy, and small intestinal hemorrhage was presumed if the patient had melena and no evidence of an upper intestinal source. Although a colonic source of blood loss may have been possible in some of these patients, NSAIDs seemed to be the cause of small intestinal bleeding.

Two NSAIDs have been reported to cause villous atrophy of the small intestine. Mefenamic acid apparently caused malabsorption, anemia, steatorrhea, and severe villous atrophy in two patients; resolution occurred after use of the drug was discontinued.24 Mefenamic acid can also cause diarrhea due to enteritis without villous atrophy.28-31 Sul-indac seemed to be the cause of diarrhea and villous atrophy in one patient.25

PREVALENCE

NSAIDs are among the most widely prescribed medications in the industrialized world; more than 100 million NSAID prescriptions are written annually by US physicians.32 Long-term NSAID therapy is common for patients with either rheumatoid arthritis or osteoarthritis.

Bjarnason and associates8 examined 97 patients with rheumatoid arthritis and osteoarthritis who had been taking NSAIDs for more than 2 months and determined that two-thirds of these patients had intestinal inflammation, as demonstrated by indium scintiscans and 4-day quantitative fecal collection of indium. Nine of these patients underwent intestinal biopsies (all peroral jejunal biopsies), one of which detected increased intraepithelial lymphocytes, a nonspecific finding.

Allison and colleagues26 determined the prevalence and morphologic features of jejunal and ileal mucosal lesions at autopsy in 713 consecutive patients with and without a history of NSAID use. The prevalence of nonspecific jejunal and ileal ulceration in patients who had taken NSAIDs was significantly increased (8.4%) in comparison with control subjects (0.6%). Patients who had received NSAIDs regularly for more than 6 months seemed to have a high risk for the development of intestinal lesions (14%). The 8.4% prevalence rate26 contrasts with the 66% prevalence rate of indium scan diagnosis of enteropathy reported by Bjarnason and coworkers.8 This difference can be explained in part by technical factors, such as the inability to detect focal microscopic intestinal erosions in the postmortem study and the lack of specificity of the radionuclide study.

In a study of 46 patients with rheumatoid arthritis who had received long-term NSAID therapy and who had iron deficiency anemia and normal findings on endoscopic examinations of the upper GI tract, total small intestinal enteroscopy was performed to determine the site of blood loss.33 In 19 of the patients, small intestinal mucosal abnormalities including erosions, ulcers, and “red spots,” which may have been early mucosal lesions, were identified. Although active bleeding was not observed, these lesions may have accounted for the blood loss.

Table 1. —Reported Cases of Nonsteroidal Anti-Inflammatory Drug-Induced Enteropathy*
Drug Indications Histopathologic and clinical findings Reference
Indomethacin AS, RA, OA, Jejunal stricture and ulcerations 9
    Felty’s syndrome Ileal stricture and ulcerations 12, 15, 22
    Subacute obstruction 17
    Perforation of terminal ileum 14,21
Piroxicam OA, RA Ileal stricture and ulcerations 11,19
    Subacute obstruction 17,23
Mefenamic acid OA Villous atrophy 16
Diclofenac sodium RA, LBP Small intestinal strictures 19
    Perforation of terminal ileum 13
Naproxen OA, RA Ileal stricture and ulcerations 15
    Thickened folds on small bowel roentgenogram 18
Azapropazone RA Subacute obstruction 17
    Jejunoileal strictures with obstruction 24
Flurbiprofen RA Jejunal stricture and ulcerations 20
Phenylbutazone OA Ileal stricture and ulcerations 10
Sulindac OA Villous atrophy 25
*AS=ankylosing spondylitis; LBP=low-back pain; OA=osteoarthritis; RA=rheumatoid arthritis.

DIAGNOSIS

In up to 50% of patients who take NSAIDs and have iron deficiency anemia, no identifiable source of blood loss can be detected by gastroscopy and colonoscopy.34 A small intestinal barium study may show strictures and areas of dilatation, factors that suggest focal obstruction even when a precise site is not evident. Small bowel strictures or diaphragms may be overlooked radiographically; thus, radiologists should be alerted to the possibility of these lesions before the study is performed.17 Nevertheless, normal findings on a small bowel barium examination do not rule out the diagnosis of NSAID-induced enteropathy.

Four radiolabeled laboratory tests have been used in assessing patients with suspected NSAID-induced enteropathy.35,36111In scintiscans of the abdomen may be used to localize areas of intestinal inflammation and are obtained at 1 to 4 hours and 20 hours after injection of the labeled leukocytes. For quantification of intestinal inflammation, an 111In quantitative fecal collection is performed. Stool specimens are collected in an inpatient setting to ensure accuracy and are collected during a 4-day period after injection of labeled leukocytes. Quantification of the marker is performed in a high-resolution bulk counter. As previously noted, two-thirds of patients receiving NSAIDs will have abnormal findings on scintiscans or increased fecal levels of 111In. The renal excretion of orally ingested 51Cr-labeled ethylenediaminetetraacetic acid can be used as a measure of intestinal permeability by quantification in urine collected 24 hours after ingestion. As an indirect assessment of NSAID-induced injury, 51Cr-labeled RBCs can be injected intravenously; stool specimens are collected for 4 days after injection to quantify intestinal blood loss. The high cost of an indium scan (more than $1,000 at many US hospitals) and the difficulty in obtaining 4-day fecal collections limit the clinical utility of these scans. To date, no study has compared a nuclear scan technique to histologic findings to assess the sensitivity and specificity of the scan for NSAID-induced enteropathy.

As previously noted, small bowel enteroscopy has been used to investigate blood loss in patients receiving NSAIDs. The advantages include direct visualization of small bowel mucosa and the ability with some prototype enteroscopes to obtain biopsy tissue for histopathologic confirmation. With current enteroscopes, however, the tip often passes rapidly around bowel loops as the instrument is withdrawn; thus, the entire small bowel mucosa is not always examined satisfactorily. In addition, the procedure can be an ordeal for patients; up to 6 hours are needed for the instrument to advance passively into the ileocecal valve in preparation for retrograde inspection of the mucosa.

Although small intestinal ulcerations and erosions are nonspecific, the presence of mucosal diaphragms is specific for intestinal damage by NSAIDs (Fig. 1). Diaphragm formation was initially described in five patients taking NSAIDs; such lesions were detected at small bowel resection.17 Because these diaphragms have morphologic features similar to prominent but otherwise normal mucosal folds, detection by barium examination is difficult. Histologically, the mucosa overlying the diaphragm shows submucosal fi-brosis with or without ulceration. The frequency of diaphragm formation in the small intestine is unknown; however, diaphragms seem to be associated with long-term (more than 1 year) NSAID use.17,18 In a retrospective study of 25 patients with NSAID-induced enteropathy, no diaphragms were found.26 In addition, diaphragmatic changes can occur in the colon, as were noted in the sigmoid colon of a patient with rheumatoid arthritis who was taking NSAIDs.37

The differential diagnosis of nonspecific erosions and ulcerations of the small intestine includes Crohn’s disease, trauma (from ingested toothpicks or bones), infection (cy-tomegalovirus, tuberculosis, or Yersinia), radiation injury, vasculitis (lupus, rheumatoid, or periarteritis nodosa), ischemia, and medications including enteric-coated potassium. Several reports have described intestinal strictures and inflammation presumably due to rheumatoid vasculitis and other collagen vascular diseases; thus, patients with such diseases should be thoroughly examined because NSAIDs are commonly used to treat these disorders.38-40

MECHANISM OF NSAID-INDUCED INJURY TO SMALL INTESTINE

Potential mechanisms of NSAID-induced injury to the small intestine are as follows.

1. NSAIDs (like aspirin) inhibit cyclooxygenase and thereby inhibit formation of prostaglandins and divert arachidonic acid metabolism to the lipoxygenase pathway.41 Prostaglandin inhibition by NSAIDs may predispose a patient to gastric ulcers and erosions.42 Whether prostaglandins are necessary to protect the small intestinal mucosa is unknown. Subcutaneous administration of indomethacin caused intestinal ulceration in rats in approximately 24 to 48 hours and gastric erosions within 3 hours.43 Maximal cyclooxygenase inhibition also occurred 3 hours after administration of indomethacin, and intestinal ulceration probably occurred by a mechanism independent of cyclooxygenase activity; however, no data are available in animals or in humans to address this issue directly.

2.NSAIDs and aspirin uncouple mitochondrial oxida- tive phosphorylation, as demonstrated in isolated animal gastric mucosa.44,45 This activity leads to a decrease in cellular adenosine triphosphate, and this depletion of high energy phosphate bonds may render the enterocyte vulnerable to damage by luminal substances. NSAIDs may also competitively inhibit enzymes in the glycolytic and Krebs’ cycle. Coadministration of glucose and citrate with indomethacin prevented the increase in intestinal permeability that occurs when only indomethacin is given to healthy volunteers, presumably because the enterocyte has been supplied with sufficient substrate to maintain normal activity of the cycles.46

Fig. 1. Endoscopic photograph of terminal ileum, demonstrating inflamed diaphragm in patient receiving long-term nonsteroidal anti-inflammatory drug therapy.

3. The enterohepatic circulation of NSAIDs may also damage the small intestine. A direct correlation exists between the concentration of NSAIDs in bile and the degree of small intestinal damage.41 Among the NSAIDs, indomethacin and piroxicam have the highest bile concentration.47 Indeed, 14 of the 25 patients with NSAID-induced enteropathy described in the literature were taking either indomethacin or piroxicam, an indication that patients may be at high risk when these two preparations are taken together (Table 1).

4. In the stomach, disruption of the mucosal barrier by NSAIDs may lead to damage of the gastric mucosa by gastric acid. In the less acidic environment of the small intestine (pH, 4.5 to 7), disruption of the intestinal barrier may lead to bacterial invasion of the intestinal wall.48 For example, indomethacin caused multiple intestinal ulcerations in rats in a standard environment; in rats that harbored a single bacterial species, intestinal ulcers developed at almost the same rate as in normal control rats. In contrast, germ-free rats were resistant to the development of intestinal lesions.49 In humans, NSAIDs also increase small intestinal permeability and allow bacterial invasion of the mucosa.50 Metronidazole seemed to decrease intestinal inflammation and blood loss as measured by fecal excretion of 111In and scans of 51Cr-labeled RBCs in patients receiving NSAIDs, although intestinal permeability changes were unaffected.51 This finding may be indirect evidence that neutrophils are attracted to invading bacteria sensitive to metronidazole.

5. Leukocyte adhesion may also have an important role in the development of NSAID-induced enteropathy. Administering a monoclonal antibody against the leukocyte-adhesion glycoprotein or antineutrophil serum was protective against naproxen- or indomethacin-induced gastric mucosal injury in rats and rabbits.52,53 Recently, aspirin was shown to promote leukocyte-endothelial cell adhesion in the mi-crovasculature of the rat mesenteric venules, a possible initiating event in NSAID-induced mucosal damage.54

TREATMENT

Data on the natural history of NSAID-induced enteropathy are scant; however, this situation may change with increasing awareness by clinicians. Recognizing the cause of the enteropathy and discontinuing use of the offending agent are the mainstay of treatment. These two factors have been reported to be successful in patients diagnosed with villous atrophy due to mefenamic acid and sulindac.16,25 Nonetheless, no data are available on the natural course of small intestinal ulcerations and diaphragms after NSAID therapy has been discontinued. In rats, pretreatment with dopamine, dimethyl sulfoxide, sucralfate, and pentagastrin may be protective against indomethacin-induced intestinal ulcera-tion.55-58 One patient with rheumatoid arthritis who had taken long-term piroxicarn therapy had multiple ileal ulcers and diaphragms. At examination, use of piroxicam was discontinued, and high-dose ornoprostil (a prostaglandin E1derivative available by prescription in Japan), sucralfate, and sulfasalazine were instituted; the intestinal bleeding and abdominal pain resolved.19 In patients who took NSAIDs for rheumatoid arthritis or osteoarthritis, misoprostol decreased intestinal permeability, as measured by urine excretion of 51Cr-labeled ethylenediamine-tetraacetic acid and L-rhamnose.59 Those doses of misoprostol were extremely high, however, and therapeutic doses of misoprostol have been unable to protect against indomethacin-induced increased intestinal permeability.60 Sulfasalazine and metronidazole may have a role in NSAID-induced enteropathy, inasmuch as both can decrease intestinal permeability.35,61

Obstruction is the most common indication for surgical treatment of NSAID-induced enteropathy. Infrequently, intestinal bleeding and perforation may also necessitate exploration. Intraoperatively, endoscopy can be performed with the endoscope placed perorally or through a surgical enterotomy. The surgeon moves the bowel over the end of the instrument—this approach allows inspection of the entire small bowel mucosa and transillumination for inspection from the serosa.

CONCLUSION

Virtually all classes of NSAIDs have been implicated in the development of enteropathy with erosions, ulcerations, blood loss, villous atrophy, and strictures. The pathogenesis is unclear but likely multifactorial. In patients taking NSAIDs who have iron deficiency and normal findings on assessment of the upper and lower GI tract, discontinuation of the NSAID therapy must be considered. Because of their lack of availability and expense, radionuclide laboratory tests have limited use. Inasmuch as NSAID-induced enteropathy is a diagnosis of exclusion, histopathologic confirmation is helpful to rule out specific causes of enteropathy. Diaphragms seem to be specific for NSAID-induced enteropathy. Thus far, no medical therapy has proved effective. Although misoprostol has been used to decrease intestinal permeability, no controlled trials have assessed effectiveness in patients with anemia. Because of the widespread use of NSAIDs, clinicians must maintain a high level of awareness for the diagnosis of NSAID-induced enteropathy.

Acknowledgment

We gratefully acknowledge Dr. Piet C. de Groen for providing the endoscopic photograph of the NSAID-related diaphragm.

REFERENCES

  1. Aabakken L.  Review article: non-steroidal, anti-inflammatory drugs—the extending scope of gastrointestinal side effects. Aliment Pharmacol Ther 1992;6:143-162
  2. Somerville K, Faulkner G, Langman M.  Non-steroidal anti-inflammatory drugs and bleeding peptic ulcer. Lancet 1986;1:462-464
  3. Armstrong CP, Blower AL.  Non-steroidal anti-inflammatory drugs and life threatening complications of peptic ulceration. Gut 1987;28:527-532
  4. Gibson GR, Whitacre EB, Ricotti CA.  Colitis induced by nonsteroidal anti-inflammatory drugs: report of four cases and review of the literature. Arch Intern Med 1992;152:625-632
  5. Tanner AR, Raghunath AS.  Colonic inflammation and non-steroidal anti-inflammatory drug administration: an assessment of the frequency of the problem. Digestion 1988;41:116-120
  6. Huber T, Ruchti C, Halter F.  Nonsteroidal antiinflammatory drug-induced colonic strictures: a case report. Gastroenterology 1991;100:1119-1122
  7. Bjarnason I, Hayllar J, MacPherson AJ, Russell AS.  Side effects of nonsteroidal anti-inflammatory drugs on the small and large intestine in humans. Gastroenterology 1993;104:1832-1847
  8. Bjarnason I, Zanelli G, Smith T, Prouse P, Williams P, Smethurst P, et al. Nonsteroidal antiinflammatory drug-induced intestinal inflammation in humans. Gastroenterology 1987;93:480-489
  9. Sturges HF, Krone CL.  Ulceration and stricture of the jejunum in a patient on long-term indomethacin therapy. Am J Gastroenterol 1973;59:162-169
  10. Neoptolemos JP, Locke TJ.  Recurrent small bowel obstruction associated with phenylbutazone. Br J Surg 1983;70:244-245
  11. Saw KC, Quick CR, Higgins AF.  Ileocaecal perforation and bleeding—are non-steroidal anti-inflammatory drugs (NSAIDs) responsible? J R Soc Med 1990;83:114-115
  12. Saverymuttu SH, Thomas A, Grundy A, Maxwell JD.  Heal stricturing after long-term indomethacin treatment. Postgrad Med J 1986;62:967-968
  13. Deakin M.  Small bowel perforation associated with an excessive dose of slow release diclofenac sodium. BMJ 1988;297:488-489
  14. Day TK.  Intestinal perforation associated with osmotic slow release indomethacin capsules. BMJ 1983;287:1671-1672
  15. Madhok R, MacKenzie JA, Lee FD, Bruckner FE, Terry TR, Sturrock RD.  Small bowel ulceration in patients receiving non-steroidal anti-inflammatory drugs for rheumatoid arthritis. QJ Med 1986;58:53-58
  16. Isaacs PE, Sladen GE, Filipe I.  Mefenamic acid enteropathy. J Clin Pathol 1987;40:1221-1227
  17. Lang J, Price AB, Levi AJ, Burke M, Gumpel JM, Bjarnason I.  Diaphragm disease: pathology of disease of the small intestine induced by non-steroidal anti-inflammatory drugs. J Clin Pathol 1988;41:516-526
  18. Bjarnason I, Price AB, Zanelli G, Smethurst P, Burke M, Gumpel JM, et al. Clinicopathological features of nonsteroidal antiinflammatory drug-induced small intestinal strictures. Gastroenterology 1988;94:1070-1074
  19. Matsuhashi N, Yamada A, Hiraishi M, Konishi T, Minota S, Saito T, et al. Multiple strictures of the small intestine after long-term nonsteroidal anti-inflammatory drug therapy. Am J Gastroenterol 1992;87:1183-1186
  20. Fellows IW, Clarke JM, Roberts PF.  Non-steroidal anti-inflammatory drug-induced jejunal and colonic diaphragm disease: a report of two cases. Gut 1992;33:1424-1426
  21. Cree IA, Walker MA, Wright M, Forrester JC.  Osmosin and ileal ulceration: a case report. Scott Med J 1985;30:40-41
  22. Laidler P, Maslin SC, Gilhome RW.  What’s new in osmosin and intestinal perforation? Pathol Res Pract 1985;180:74-76
  23. Sukumar L.  Recurrent small bowel obstruction associated with piroxicam. Br J Surg 1987;74:186
  24. Johnston F.  Recurrent small bowel obstruction associated with piroxicam [letter]. Br J Surg 1987;74:654
  25. Freeman HJ.  Sulindac-associated small bowel lesion. J Clin Gastroenterol 1986;8:569-571
  26. Allison MC, Howatson AG, Torrance CJ, Lee FD, Russell RI.  Gastrointestinal damage associated with the use of nonsteroidal antiinflammatory drugs. N Engl J Med 1992;327:749-754
  27. Langman MJ, Morgan L, Worrall A.  Use of anti-inflammatory drugs by patients admitted with small or large bowel perforations and haemorrhage. BMJ 1985;290:347-349
  28. Hall RI, Petty AH, Cobden I, Lendrum R.  Enteritis and colitis associated with mefenamic acid. BMJ 1983;287:1182
  29. Rampton DS, Tapping PJ.  Enteritis and colitis associated with mefenamic acid [letter]. BMJ 1983;287:1627
  30. Williams R, Glazer G.  Enteritis and colitis associated with mefenamic acid [letter]. BMJ 1983;287:1627
  31. Phillips MS, Fehilly B, Stewart S, Dronfield MW.  Enteritis and colitis associated with mefenamic acid [letter]. BMJ 1983;287:1626
  32. Gibson T.  Nonsteroidal anti-inflammatory drugs—another look. Br J Rheumatol 1988;27:87-90
  33. Morris AJ, Wasson LA, MacKenzie JF.  Small bowel enteroscopy in undiagnosed gastrointestinal blood loss. Gut 1992;33:887-889
  34. Upadhyay R, Torley HI, McKinlay AW, Sturrock RD, Russell RI.  Iron deficiency anaemia in patients with rheumatic disease receiving non-steroidal anti-inflammatory drugs: the role of upper gastrointestinal lesions. Ann Rheum Dis 1990;49:359-362
  35. Bjarnason I, Zanelli G, Prouse P, Smethurst P, Smith T, Levi S, et al. Blood and protein loss via small-intestinal inflammation induced by non-steroidal anti-inflammatory drugs. Lancet 1987;2:711-714
  36. Bjarnason I, Williams P, So A, Zanelli GD, Levi AJ, Gumpel JM, et al. Intestinal permeability and inflammation in rheumatoid arthritis: effects of non-steroidal anti-inflammatory drugs. Lancet 1984;2:1171-1174
  37. Hershfield NB.  Endoscopic description of “diaphragm disease” induced by non-steroidal anti-inflammatory drugs [abstract]. Gastrointest Endosc 1992;38:267
  38. Kuehne SE, Gauvin GP, Shortsleeve MJ.  Small bowel stricture caused by rheumatoid vasculitis. Radiology 1992;184:215-216
  39. Dougados M, Alemanni M, Tulliez M, Abadia R, Kahan A, Delrieu F, et al. Systematic ileocolonoscopy in seronegative spondylarthropathies. Rev Rhum Mai Osteoartic 1987;54:279-283
  40. Okuda Y, Takasugi K, Imai A, Kondo Y, Hachinota M, Ueda S, et al. Two cases of rheumatoid arthritis complicated with vasculitis-induced ischemic enterocolitis. Ryumachi 1990;30:403-407
  41. Rainsford KD.  Mechanisms of gastrointestinal toxicity of non-steroidal anti-inflammatory drugs. Scand J Gastroenterol Suppl 1989;163:9-16
  42. Soll AH, Weinstein WM, Kurata J, McCarthy D.  Nonsteroidal anti-inflammatory drugs and peptic ulcer disease. Ann Intern Med 1991;114:307-319
  43. Whittle BJ.  Temporal relationship between cyclooxygenase inhibition, as measured by prostacyclin biosynthesis, and the gastrointestinal damage induced by indomethacin in the rat. Gastroenterology 1981;80:94-98
  44. Jorgensen TG, Weis-Fogh US, Neilsen HH, Olesen HP.  Salicylate- and aspirin-induced uncoupling of oxidative phosphorylation in mitochondria isolated from the mucosal membrane of the stomach. Scand J Clin Lab Invest 1976;36:649-654
  45. Spenney JG, Bhown M.  Effect of acetylsalicylic acid on gastric mucosa. II. Mucosal ATP and phosphocreatine content, and salicylate effects on mitochondrial metabolism. Gastroenterology 1977;73:995-999
  46. Bjarnason I, Smethurst P, Macpherson A, Walker F, McElnay JC, Passmore AP, et al. Glucose and citrate reduce the permeability changes caused by indomethacin in humans. Gastroenterology 1992;102:1546-1550
  47. Beck WS, Schneider HT, Dietzel K, Nuernberg B, Brune K.  Gastrointestinal ulcerations induced by anti-inflammatory drugs in rats: physicochemical and biochemical factors involved. Arch Toxicol 1990;64:210-217
  48. Wilson TH.  Intestinal Absorption. Philadelphia: Saunders, 1962: 130
  49. Robert A, Asano T.  Resistance of germfree rats to indomethacin-induced intestinal lesions. Prostaglandins 1977;14:333-341
  50. Bjarnason I, Williams P, Smethurst P, Peters TJ, Levi AJ.  Effect of non-steroidal anti-inflammatory drugs and prostaglandins on the permeability of the human small intestine. Gut 1986;27:1292-1297
  51. Bjarnason I, Hayllar J, Smethurst P, Price A, Gumpel MJ.  Metronidazole reduces intestinal inflammation and blood loss in non-steroidal anti-inflammatory drug induced enteropathy. Gut 1992;33:1204-1208
  52. Wallace JL, Keenan CM, Granger DN.  Gastric ulceration induced by nonsteroidal anti-inflammatory drugs is a neutro-phil-dependent process. Am J Physiol 1990;259:G462-G467
  53. Wallace JL, Arfors KE, McKnight GW.  A monoclonal antibody against the CD 18 leukocyte adhesion molecule prevents indomethacin-induced gastric damage in the rabbit. Gastroenterology 1991;100:878-883
  54. Asako H, Kubes P, Wallace J, Wolf RE, Granger DN.  Modulation of leukocyte adhesion in rat mesenteric venules by aspirin and salicylate. Gastroenterology 1992;103:146-152
  55. Sikiric P, Rotkvic I, Mise S, Krizanac S, Gjuris V, Jukic J, et al. The influence of dopamine agonists and antagonists on indomethacin lesions in stomach and small intestine in rats. Eur J Pharmacol 1988;158:61-67
  56. Ali BH.  Protective effect of dimethylsulphoxide in stress- and indometacin-induced gastrointestinal ulceration in rats. Pharmacology 1989;39:98-102
  57. Waisman Y, Zahavi I, Marcus H, Ligumsky M, Rosenbach Y, Dinari G.  Sucralfate is protective against indomethacin-induced intestinal ulceration in the rat. Digestion 1988;41:78-82
  58. Harel L, Zahavi I, Marcus H, Waisman Y, Ligumsky M, Rosenbach Y, et al. Pentagastrin protects the proximal small intestine against indomethacin-induced ulcers in the rat. Digestion 1987;38:156-159
  59. Bjarnason I.  Experimental evidence of the benefit of misoprostol beyond the stomach in humans. J Rheumatol Suppl 1990;20:38-41
  60. Davies GR, Rampton DS.  Relevance of intestinal bacteria, prostaglandins, and pro-drug administration to non-steroidal anti-inflammatory drug (NSAID) induced increases in gut permeability [abstract]. Gut 1991;32:A1224
  61. Bjarnason I, Hopkinson N, Zanelli G, Prouse P, Smethurst P, Gumpel JM, et al. Treatment of non-steroidal anti-inflammatory drug induced enteropathy. Gut 1990;31:777-780